D prostanoid receptor 2 (chemoattractant receptor–homologous molecule expressed on TH2 cells) protein expression in asthmatic patients and its effects on bronchial epithelial cells

Background The D prostanoid receptor 2 (DP2; also known as chemoattractant receptor–homologous molecule expressed on TH2 cells) is implicated in the pathogenesis of asthma, but its expression within bronchial biopsy specimens is unknown. Objectives We sought to investigate the bronchial submucosal DP2 expression in asthmatic patients and healthy control subjects and to explore its functional role in epithelial cells. Methods DP2 protein expression was assessed in bronchial biopsy specimens from asthmatic patients (n = 22) and healthy control subjects (n = 10) by using immunohistochemistry and in primary epithelial cells by using flow cytometry, immunofluorescence, and quantitative RT-PCR. The effects of the selective DP2 agonist 13, 14-dihydro-15-keto prostaglandin D2 on epithelial cell migration and differentiation were determined. Results Numbers of submucosal DP2+ cells were increased in asthmatic patients compared with those in healthy control subjects (mean [SEM]: 78 [5] vs 22 [3]/mm2 submucosa, P < .001). The bronchial epithelium expressed DP2, but its expression was decreased in asthmatic patients compared with that seen in healthy control subjects (mean [SEM]: 21 [3] vs 72 [11]/10 mm2 epithelial area, P = .001), with similar differences observed in vitro by primary epithelial cells. Squamous metaplasia of the bronchial epithelium was increased in asthmatic patients and related to decreased DP2 expression (rs = 0.69, P < .001). 13, 14-Dihydro-15-keto prostaglandin D2 promoted epithelial cell migration and at air-liquid interface cultures increased the number of MUC5AC+ and involucrin-positive cells, which were blocked with the DP2-selective antagonist AZD6430. Conclusions DP2 is expressed by the bronchial epithelium, and its activation drives epithelial differentiation, suggesting that in addition to its well-characterized role in inflammatory cell migration, DP2 might contribute to airway remodeling in asthmatic patients.

D prostanoid receptor 2 (DP2) or chemoattractant receptorhomologous molecule expressed on T H 2 cells (CRTH2) is a G protein-coupled receptor that has been implicated in the pathogenesis of allergic diseases. 1 DP2 is activated by prostaglandin D 2 (PGD 2 ), which is found at high levels in the bronchoalveolar lavage fluid of asthmatic patients. [2][3][4][5][6] The expression of DP2 and the effects of its stimulation on T H 2 lymphocytes, eosinophil and basophil migration, and activation has been well characterized. 7,8 An increase in the number of DP2 1 inflammatory cells in patients with allergic disease 6 has highlighted a potential role for this receptor in allergy 9,10 and asthma. 11,12 In addition, considerable interest in the development of DP2 antagonists in both patients with allergic conditions 9,10 and asthmatic patients [11][12][13] has strengthened the linkage between the DP2 receptor and inflammatory-related disorders. To date, limited efficacy has been demonstrated for DP2 antagonists in asthmatic patients 11,13 ; however, it remains to be determined whether DP2 antagonism is more effective in a subset of patients. A thorough understanding of DP2 expression within the airways and whether changes in receptor expression correlates with disease severity might aid in identifying a responsive asthmatic group. Unfortunately, there is currently a lack of data describing the protein expression of DP2 in bronchial biopsy specimens in asthmatic patients, which has limited the cell types that DP2 function has been explored within. In contrast, the expression of DP2 on epithelial cells has been described from a variety of tissues, including the nose, 14 skin, 15 and retina, 16 and bronchial epithelial cells in patients with chronic obstructive pulmonary disease (COPD). 17 These studies highlight the potential that DP2 might be expressed on epithelial cells within the airways of asthmatic patients. In addition, in mouse challenge models 18,19 DP2 antagonists caused a reduction in goblet cell hyperplasia, suggesting that DP2 activation on epithelial cells might play a key role in the pathogenesis of asthma. Comparative studies looking at the expression of DP2 on biopsy specimens from asthmatic patients and healthy control subjects would provide useful data to indicate possible target cells located within the airways for DP2 antagonists, helping to focus future DP2 antagonist study readouts and patient populations. We hypothesized (1) that submucosal inflammatory cells and the bronchial epithelium express DP2 and that its expression is increased in asthmatic patients and related to disease severity and (2) that activation of DP2 in primary epithelial cells promotes migration and differentiation. To test our hypotheses, we have undertaken an immunohistochemical analysis of bronchial biopsy specimens from asthmatic patients and healthy control subjects and studied the expression and function of DP2 in primary epithelial cells in submerged and air-liquid interface (ALI) cultures.

METHODS Subjects
Healthy control subjects and asthmatic patients were recruited from Glenfield Hospital, Leicester, United Kingdom. Asthma severity was defined according to the Global Initiative for Asthma treatment steps. 20 Subjects were characterized in terms of demographics, smoking history, spirometry, sputum cell counts, and atopic status, which was defined as either 1 or more positive skin prick test responses or blood-specific IgE levels to common aeroallergens. Healthy subjects had no history of respiratory or allergic disease and had normal spirometric results. The study was approved by the Leicestershire Research Ethics Committee. Informed consent was obtained from all subjects.

Epithelial cell expression
Extracellular and intracellular (0.1% saponin) DP2 expression was assessed with DP2-PE antibody relative to isotype control (Rat-PE isotype) with the FACSAria (BD Biosciences, Oxford, United Kingdom). The effects of corticosteroids on DP2 expression were investigated, as described above, by incubating healthy control cells with 1 mmol/L fluticasone propionate (Sigma) for 24 hours. The effects of 100 nmol/L DK-PGD 2 for 24 hours on DP2 expression were also assessed. For the study of DP2 expression in submerged cells grown in chamber slides, cells were fixed, stained with AstraZeneca DP2 antibody, and detected with anti-rabbit Alexa Fluor 488 (Invitrogen, Paisley, United Kingdom).

Cell migration
The Oris Cell Migration Kit was used (tebu-bio, Peterborough, United Kingdom). Triplicate repeats for vehicle control (1 mmol/L dimethyl sulfoxide), 100 nmol/L DK-PGD 2 or 100 nmol/L DK-PGD 2 , and 1 mmol/L AZD6430 were added for 24 hours in 5 healthy control donors and 5 asthmatic donors. The concentrations of 500 nmol/L and 1 mmol/L DK-PGD 2 were tested in cells from 5 healthy control donors. TGF-b1 (10 ng/mL) and fibroblast growth factor (25 ng/mL; R&D Systems) were used as positive controls in cells from 5 healthy control donors and 2 asthmatic donors. Cells were fixed and labeled with Hoechst nuclear dye (Invitrogen, Carlsbad, Calif). The number of cells migrated into the migration zone was counted by a blinded observer.

Analysis
Statistical analysis was performed with PRISM software, version 6 (GraphPad Software, La Jolla, Calif). Parametric data were analyzed with 1-or 2-way ANOVA, Tukey posttest correction for intergroup comparison, or the paired t test. Nonparametric data were analyzed with the Kruskal-Wallis test and the Dunn test for post hoc comparison. The Spearman correlation test was used for correlation analysis. A P value of less than .05 was considered significant.

Immunohistochemistry staining for DP2 on biopsy specimens
Clinical characteristics of the patients with mild, moderate, or severe asthma and healthy control subjects are shown in Table I. Groups were well matched for age and smoking history. Asthmatic patients had impaired lung function and evidence of eosinophilic airway inflammation. Representative examples of DP2 expression in bronchial biopsy specimens from asthmatic patients and healthy control subjects are shown (Fig 1, A-D). No staining was seen for the isotype control (Fig 1, A) or when the antibody was incubated with a blocking peptide (see Fig E1 in this article's Online Repository at www.jacionline.org). Expression with a commercially available DP2 antibody was similar (see Figs E2 and E3 in this article's Online Repository at www.jacionline.org).
DP2 expression was observed on inflammatory cells within the submucosa for biopsy specimens from asthmatic patients and those from healthy control subjects. Quantification of DP2 1 cells within the submucosa demonstrated a significant increase in biopsy specimens from patients with severe asthma compared with that seen in biopsy specimens from healthy control subjects (mean [SEM]: 78 [5] vs 22 [3] cells/mm 2 submucosa, P < .001; Fig 1, E). By using serial section staining, DP2 was found to colocalize with a subset of CD3 1 T cells, major basic proteinpositive eosinophils, and tryptase-positive mast cells (Fig 1, F). DP2 1 submucosal cells were most commonly T cells. The number of DP2 1 CD3 1 cells was significantly increased in biopsy specimens from patients with mild, moderate, and severe asthma compared with that seen in specimens from healthy control subjects (mean [SEM]: 25 [5]  submucosa, P 5 .030). In a subset of asthmatic patients (n 5 12) and healthy control subjects (n 5 5), the DP2 1 CD3 1 phenotype was investigated further by using CD4 and CD8 markers. The number of DP2 1 CD4 1 cells was significantly increased in asthmatic patients compared with that seen in healthy control subjects (mean [SEM]: 15 [3] vs 4 [1] mm 2 submucosa, P 5 .002). DP2 expression was also observed for CD8 1 cells, but only a small proportion of cells and no significant differences between biopsy specimens from healthy control subjects and asthmatic patients were found (data not shown). There was no colocalization of DP2 1 cells with neutrophils. Positive expression was seen for biopsy specimens from asthmatic patients and healthy control subjects on epithelial cells. The number of DP2 1 epithelial cells was significantly reduced in biopsy specimens from patients with moderate and severe asthma compared with those from healthy control subjects (Fig 1, G; mean [SEM]: 30 [5] vs 72 [11]/10 mm 2 epithelium, P 5 .036; 21 [3] vs 72 [11]/10 mm 2 epithelium, P 5 .001) and in biopsy specimens from patients with severe asthma compared with those from patients with mild asthma (mean [SEM]: 21 [3] vs 54 [8]/10 mm 2 epithelium, P 5 .027). The number of DP2 1 inflammatory cells and DP2 1 epithelial cells had reciprocal correlations with total sputum cell counts (r 5 0.54, P 5.003; r 5 20.42, P 5.028), FEV 1 percentage bronchodilator response (r 5 0.36, P 5 .048; r 5 20.43, P 5 .002), and airway hyperresponsiveness (r 5 20.586, P 5 .004; r 5 0.55, P 5 .009), respectively, but not with FEV 1 percent predicted, FEV 1 /forced vital capacity ratio, or sputum differential cell counts.
To determine whether a change in phenotype had occurred in the DP2 2 epithelial cells, we costained cells with involucrin, which was previously described as a reliable marker of a squamous metaplastic phenotype by Araya et al 23 in the lungs of patients with COPD. We found a lack of DP2 staining on epithelial cells in areas expressing pancytokeratin (used to confirm epithelial origin) and involucrin (Fig 2, A-C). Many of the DP2 2 epithelial cells had a flattened squamous morphology (Fig 2, C). These findings suggested that the reduction in DP2 1 epithelial cell counts was due to a metaplastic change in phenotype of the epithelial cells in the groups with moderate and severe asthma. Epithelial histology for all biopsy specimens were graded according to the criteria shown   (Fig 1, A), a healthy control subject (Fig 1,  B), a patient with mild asthma (Fig 1, C), and a patient with severe asthma (Fig 1, D).

DP2 expression on cultured epithelial cells
To investigate whether differences in DP2 expression in vivo also existed in vitro, we characterized the expression of DP2 on cultured epithelial cells taken from healthy subject and asthmatic patients. All asthmatic donors used had moderate-to-severe disease (Global Initiative for Asthma treatment steps 3-5). Fluorescent DP2 cell staining was associated with submerged epithelial cells from healthy subjects and asthmatic patients (Fig 3, A and  B). For epithelial cells from asthmatic patients, there were some cells that were DP2 2 (49-6-diamidino-2-pheynylindole dihydrochloride-positive but DP2 2 ; Fig 3, B). No DP2 staining was observed with the isotype control (Fig 3, A, insert). DP2 expression was present on epithelial cells grown in an ALI format from both healthy (Fig 3, C) and asthmatic (Fig 3, D) Fig 2, A), pancytokeratin (Fig 2, B), and DP2 (Fig 2, C). D, Epithelial histology grades for biopsy specimens from healthy control subjects and patients with mild, moderate, and severe asthma.  A, Green staining for DP2, with blue 49-6-diamidino-2-pheynylindole dihydrochloride (DAPI) nuclear staining (cells from healthy control subjects). The inset shows a rabbit isotype control with lack of any green staining. B, Green staining for DP2, with blue DAPI nuclear staining (cells from asthmatic patients). Note cells with absence of DP2 1 cells (green) staining. C, Green staining for DP2 on ALI culture from healthy control subjects. D, Green staining for DP2 on ALI culture from asthmatic patients.   Fig 3, F). Because we did not detect any DP1 mRNA expression on epithelial cells grown in submerged or ALI culture, the role of DP1 was not investigated (data not shown).

DP2 activation modulates epithelial differentiation in ALI cultures
The linkage of DP2 expression with the epithelial phenotype within biopsy specimens led us to hypothesize that DP2 activation might play a role in epithelial differentiation. To investigate this, ALI cultures were used because they contain cells in variable states of differentiation. Five separate donors of healthy ALI cultures were treated with vehicle control, DK-PGD 2 , or DK-PGD 2 and AZD6430. Treatment for 24 hours with DK-PGD 2 produced an increase in goblet cell numbers quantified by using MUC5AC 1 staining, which could be blocked with AZD6430 (Fig 4, A, C, and F). No differences were observed between untreated and vehicle control-treated cultures. A significant fold increase in the number of MUC5AC 1 cells was observed with DK-PGD 2 treatment when compared with untreated ALI (4-fold [3-to 4-fold] increase, P < _ .001), which decreased significantly in the presence of AZD6430 (Fig 4, B, C, and F). IL-13, which was used as a positive control, caused a significant fold increase in the number of MUC5AC 1 cells compared with untreated values (IL-13: 5-fold [4-to 7-fold] increase, P < .001; Fig 4, D and F). AZD6430 did not affect IL-13 responses (5.6fold [4-to 7-fold] vs 5-fold [4-to 7-fold] with IL-13 alone; Fig  4, D and E). MUC5AC mRNA analysis showed similar results to the protein expression (Fig 4, G). More chronic effects of DK-PGD 2 incubation were also assessed at 48 and 72 hours. An increase in MUC5AC 1 cell numbers compared with untreated values was maintained at 48 hours of DK-PGD 2 (2-fold [1to 4-fold], P 5 .021), but this effect diminished at 72 hours (Fig 4, H).

DISCUSSION
Here we present compelling evidence that DP2 is differentially expressed on inflammatory and epithelial cells in the airways of patients with moderate-to-severe asthma when compared with those of healthy control subjects. More importantly, accumulation of DP2 1 T cells in the bronchial submucosa was closely associated with asthma severity. We also show that DP2 activation in epithelial cells induces proremodeling responses. These findings demonstrate that activation of DP2 in T cells and the epithelium has the potential to drive key features of severe asthma.
Few studies have investigated the expression of DP2 in the airways of patients with lung diseases and how this expression correlates with disease severity. Previous studies have shown increased numbers of DP2 1 T cells in the nasal mucosa of allergic compared with nonallergic subjects 14 and in the bronchoalveolar lavage fluid cells of patients with severe asthma compared with numbers seen in healthy subjects. 6,24 Our study is the first to demonstrate an increased infiltration of DP2 1 T cells in the bronchial submucosa of patients with moderate-tosevere asthma when compared with values in healthy subjects. Further analysis of the DP2 1 T cells on a subset of biopsy specimens demonstrated that the majority of T cells were CD4 1 , although a small proportion of CD8 1 T cells were also DP2 1 . In addition, we speculate that type 2 innate lymphoid cells contribute to the total number of DP2 1 T cells because these have been previously been found to express DP2. 25,26 DP2 has been found to delay apoptosis of T H 2 lymphocytes, 27 and the findings from our study support the concept that this action might cause T cells to be retained within the submucosa of the airways. Increased numbers of DP2 1 eosinophils were found in biopsy specimens from patients with moderate asthma. DP2 antagonists have been found to reduce sputum eosinophil numbers in allergen-challenged steroid-naive asthmatic patients 12 and nasal eosinophil numbers in patients with allergic rhinitis. 9 Our data suggest that there is a potential that DP2 antagonists can affect tissue eosinophil numbers in patients with moderate asthma. Although DP2 was also found on mast cells (with a lack of expression on neutrophils), there was no significant difference between healthy subjects and asthmatic patients in DP2 expression on these cell types. The contribution of DP2 in the  (Fig 4, A) and treatment with 100 nmol/L DK-PGD 2 for 24 hours (Fig 4, B), and 100 nmol/L DK-PGD 2 plus 1 mmol/L AZD6430 for 24 hours (Fig 4, C). D, IL-13 (100 ng/mL) for 24 hours. E, IL-3 (100 ng/mL) plus AZD6430 (1 mmol/L) for 24 hours. pathogenesis of asthma has not been completely elucidated. Activation of DP2 on T H 2 cells has been shown to cause an increase in the ability of these cells to produce IL-2, IL-4, IL-5, and IL-13. 28,29 In turn, these cytokines could regulate key features of severe asthma, as suggested by preclinical studies showing that DP2 antagonism significantly reduced allergeninduced inflammatory changes within mouse airways. 18 Therefore activation of accumulated DP2 1 T cells within the airways of asthmatic patients is likely to play a significant role in the pathogenesis of allergic asthma through proinflammatory actions.
There is a wealth of literature suggesting that epithelial cells contribute to remodeling changes within the airways of asthmatic patients (as reviewed by Davies 30 ). Epithelial cells are more stressed in asthmatic patients, showing upregulation of activated transcription factors, 31 and activated repair processes are evidenced by increased epithelial growth factor receptor 32,33 and a persistently defective barrier. 34 We found that DP2 was expressed on epithelial cells within biopsy specimens from asthmatic patients. Although previous publications have described the expression of DP2 on cultured normal human epithelial cells and H292 cells, 35 our report is the first to demonstrate the in vivo expression of DP2 on epithelial cells within bronchial biopsy specimens. A recent report has described expression of DP2 on epithelial cells in lung volume reduction tissue from patients with COPD, 17 but whether this FIG 5. A-E, Representative images of ALI cultures from healthy control subjects of involucrin-positive staining (brown, 3400 magnification). Results are shown for untreated conditions (Fig 5, A) and treatment with 100 nmol/L DK-PGD 2 for 48 hours (Fig 5, B), and 100 nmol/L DK-PGD 2 plus 1 mmol/L AZD6430 for 48 hours (Fig 5, C). D, TGF-b1 (10 ng/mL) for 72 hours. E, TGF-b1 (10 ng/mL) plus AZD6430 (1 mmol/L) for 72 hours. expression was associated with disease severity was not investigated because no healthy control tissue was included. We have shown that the numbers of DP2 1 epithelial cells were significantly decreased in the airway epithelium in patients with moderate-to-severe asthma. Further investigations led us to demonstrate that the epithelial cell phenotype in biopsy specimens of patients with moderate-to-severe asthma was dramatically altered when compared with that in healthy control subjects. We found that there were frequent areas of squamous metaplasia in patients with moderate-to-severe asthma when using the involucrin marker previously described in patients with COPD. 23 Interestingly, quantification of epithelial changes to those seen with a metaplastic phenotype inversely correlated with DP2 expression. Squamous metaplasia and mucous cell metaplasia are the most common metaplastic features associated with epithelial tissue. 36 More importantly, squamous metaplasia has been found to correlate with the severity of airway obstruction 37 and to increase with the severity of COPD, 23 possibly related to cigarette smoke exposure because squamous metaplasia is more frequent in asthmatic patients who smoke. 38 In this study squamous metaplasia was increased in the moderate-to-severe asthma cohort without a difference in smoking status between the groups. This suggests that there might be other factors independent of cigarette smoke which can contribute to the induction of squamous metaplasia in airway epithelium. Our data indicate that in patients with moderateto-severe asthma, a phenotype shift of epithelial cells can occur, which influences DP2 expression. Interestingly, differential expression between healthy control subjects and patients with moderate-to-severe asthma was also maintained for epithelial cells when grown in culture. This finding could suggest that cultured epithelial cells from asthmatic patients have an intrinsically altered phenotype, an observation that has been suggested in previous studies using epithelial cells from children with asthma. 39 In our study DP2 was found to have intracellular and extracellular expression on epithelial cells similar to that seen in previous studies for this receptor. 14,35 DP2 activation caused functional consequences on epithelial cells that were likely mediated through cell-surface receptors. However, intracellular receptor activation can also occur, as has been reported for other G protein-coupled receptors, 40 and its functional importance requires further study.
The novel observation of DP2 expression on bronchial epithelial cells directed further investigation into the functional effects of DP2 activation on cultured bronchial epithelial cells. DP2 activation through PGD 2 has been shown to cause cell migration in T H 2 cells, basophils, and eosinophils. 8,41 In the current study we found that DP2 activation with the DP2-selective agonist DK-PGD 2 42,43 also caused migration of cells from both asthmatic patients and healthy subjects, an effect that was blocked with a DP2-selective antagonist. However, although the antagonist was highly selective, it had low affinity for other receptors and enzymes, such as thromboxane receptor, and thus we cannot fully exclude off-target effects. Migration was more pronounced in cells from healthy control subjects, possibly because of the difference in cell-surface receptor expression. We recognize that within this study, a dose-response curve was not fully explored in part because of limitations of cell numbers, but increasing concentrations of DK-PGD 2 up to 1 mmol/L revealed that maximal migratory responses were obtained. Future studies comparing different concentrations of DK-PGD 2 with more potent DP2 agonists, such as 15(R)-15methyl-PGD 2 , might help in uncovering differences between the asthmatic and healthy states. The existence of functional DP2 on the epithelium is supported with work in mouse models, where DP2 antagonists have been found to influence mucous cell metaplasia and epithelial cell hyperplasia in response to cigarette smoke 19 or allergen stimulation. 44 Using the ALI culture system, which closely mimics the in vivo environment, 45 we provide additional evidence for a role of DP2 in driving phenotype changes of the epithelial cells by showing that DK-PGD 2 treatment induced not only increased goblet cell numbers when exposed acutely but also increased the area of involucrin expression in the epithelium with more chronic treatment. The positive controls used in this study validated these responses, in which IL-13 significantly upregulated the number of MUC5AC 1 cells 46 and TGF-b1 significantly increased the amount of involucrin expression. 47 Epithelial cells are thought to be highly plastic in that they can rapidly change their phenotype in response to insult. 48 The classical repair response of epithelial cells to injury is thought to consist of a number of steps. These include transient mucus release, shedding of columnar epithelial cells, spreading and migration of basal epithelial cells, and induction of squamous metaplasia through progressive redifferentiation, ultimately leading to regeneration of the mucociliary epithelium. [48][49][50] The expression of DP2 on basal and columnar epithelial cells and the findings that DP2 activation can cause many of these repair step processes could indicate that this receptor plays a key role in the maintenance and repair of the epithelial barrier. In asthmatic patients, in whom there is an increased presence of PGD 2 , 2,3,6 it is likely that DP2 activation accelerates these functional responses on epithelial cells, causing an aberrant mucosal barrier phenotype. Such changes can aid progression of disease and make patients more susceptible to respiratory tract infection. [51][52][53] Therefore a DP2 antagonist might be useful in decreasing DP2 activation on epithelial cells and restoring normal epithelial differential processes.
In conclusion, we have described the differential expression of DP2 on biopsy specimens from healthy control subjects and asthmatic patients. Biopsy specimens from patients with severe asthma were associated with increased DP2 1 T-cell numbers within the submucosal compartment and reduced DP2 1 epithelial cell numbers in areas of epithelial metaplasia. Some of the epithelial features seen in patients with severe asthma could be reproduced by activating DP2 on bronchial epithelial cells, causing cell migration and an increase in numbers of goblet cells and cells of a squamous phenotype. The effects of DP2 activation on epithelial cells might influence airway remodeling processes in asthmatic patients, and end points, such as mucus production, should be considered in future clinical DP2 antagonist studies. We conclude that a DP2 antagonist might not just inhibit infiltration of DP2 1 inflammatory cells into the airways but might also act on epithelial cells and prevent proremodeling action.
We thank Iain Dougall for his assistance with manuscript review. We also thank Hilary Marshall for her assistance with processing of the ALI cultures. Dr John Mo from AstraZeneca was involved in the approval of the manuscript submission.
Key messages d DP2 is differentially expressed on inflammatory cells and bronchial epithelial cells in biopsy specimens from patients with asthma compared with those from healthy control subjects.
d DP2 activation on bronchial epithelial cells might contribute to airway remodeling in asthmatic patients.